Claude Code Security is an AI-powered security solution integrated into Claude Code that helps organizations proactively defend their software from vulnerabilities. Unlike traditional static analysis tools that rely on predefined rules, it reasons through code the way a human security researcher would. By understanding business logic, tracing data flows, and examining component interactions, it detects subtle and high-severity vulnerabilities that automated scanners often miss. Every identified issue passes through a layered self-verification process in which the AI attempts to confirm or refute its own findings to minimize false positives. The system then assigns severity and confidence ratings so teams can focus on the most urgent threats. Within the security dashboard, developers can review detailed explanations and inspect AI-generated patch suggestions before making any changes. Human oversight remains central, as no fixes are applied automatically without approval. Built on Claude Opus 4.6, the technology has already uncovered hundreds of long-hidden vulnerabilities in open-source projects. The tool is being released as a limited research preview to Enterprise and Team customers, with expedited access for open-source maintainers. By equipping defenders with advanced AI-driven analysis, Claude Code Security aims to raise the overall security baseline across the software industry.
